---
title: "Datenauswertung TxKohorte Mai 2018"
#author: "Stephan Glöckner"
email: "stephan.gloeckner@helmholtz-hzi.de"
output: 
  flexdashboard::flex_dashboard:
    orientation: rows
    vertical_layout: fill
    logo:  pics/hzi_logo.png
    css: ["css/flex.css", "css/summarytools.css"]
    source_code: embed
    navbar:
      - { title: "Back to Talk", href: "index.html#10"}
---

```{r setup, include=FALSE}
library(tidyverse)
library(lubridate)
library(plotly)
library(flexdashboard)
library(kableExtra)
library(DT)
library(knitr)
library(tidyverse)
library(viridis)
library(formattable)
library(RColorBrewer)

```

Studienpopulation 
=======================================================================
```{r pop_data_gen, echo=FALSE, warning=FALSE, message=FALSE}
# data simulation - study population
dfr <- tibble(Site = sample(paste0("Center_", 1:5), 2000, replace = TRUE),
             Organ = c(sample("Heart", 58, replace = TRUE),
                       sample("Liver", 89, replace = TRUE),
                       sample("Lung", 47, replace = TRUE),
                       sample("Kidney", 871, replace = TRUE),
                       sample("Pancreas", 589, replace = TRUE),
                       sample("Stem Cells", 346, replace = TRUE)),
             date = sample(seq(as.Date('2017/01/01'), as.Date('2018/08/31'), by="day"), 2000, replace = TRUE)
             )
# tab_overview
tab_overview <- dfr %>% 
  count(Site, Organ) %>% 
  spread(Organ, n) %>% 
  bind_rows(., tibble(Site = "TOTAL", 
                      Heart = sum(.$Heart, na.rm=T),
                      Kidney = sum(.$Kidney, na.rm=T),
                      Liver = sum(.$Liver, na.rm=T),
                      Lung = sum(.$Lung, na.rm=T),
                      Pancreas = sum(.$Pancreas, na.rm=T),
                      `Stem Cells` = sum(.$`Stem Cells`, na.rm=T))) %>% 
  mutate(TOTAL = rowSums(.[2:7])) 

# fig_overview
fig_overview <- tab_overview %>% 
  gather(key, value, -Site) %>% 
  filter(key != "TOTAL",
         Site != "TOTAL") %>% 
  plotly::plot_ly(x = ~value, y = ~Site, color = ~key, colors = viridis_pal(option = "D")(6)) %>%
  layout(barmode='stack')

# graph_site
graph_site <- dfr %>%
  select(Site, date) %>%
  count(Site, date) %>% 
  group_by(Site) %>%
  mutate(cumsum=cumsum(n)) %>%
  select(-n) %>%
  plotly::plot_ly(
    x = ~date, 
    y = ~cumsum,
    color = ~Site,
    colors = viridis_pal(option = "D")(5)
  ) %>%
  add_lines() %>%
  layout(
    xaxis = list(title = "Date",zeroline = F),
    yaxis = list(title = "Patienten", zeroline = F)
  )

```



Row 
-----------------------------------------------------------------------

### Grafik Übersicht Studienpopulation
```{r}
fig_overview
```

Row 
-----------------------------------------------------------------------
### Populationstrend nach Zentrum
```{r}
graph_site
```


### Tabelle Übersicht Studienpopulation
```{r}
tab_overview %>% 
  knitr::kable() %>%
  kable_styling()
```

### Auf einem Blick
```{r}

```


Bioproben {data-orientation=rows}
=======================================================================

```{r bio_data_gen, echo=FALSE, warning=FALSE, message=FALSE}


# biosample generation
blood_types <- c("EDTA-Plasma", "EDTA-Primärröhrchen", "PBMCs", "RNA", "Serum", "keine Angabe", NA, "EDTA-Buffy Coat")
other_types <- c("Stuhl", "Urin", NA, "Bukkaler Abstrich", "Rachenspülwasser")

dfr_bio <- tibble(Site = sample(paste0("Center_", 1:5), 2000*7, replace = TRUE),
       Organ = c(sample("Heart", 58*7, replace = TRUE),
                 sample("Liver", 89*7, replace = TRUE),
                 sample("Lung", 47*7, replace = TRUE),
                 sample("Kidney", 871*7, replace = TRUE),
                 sample("Pancreas", 589*7, replace = TRUE),
                 sample("Stem Cells", 346*7, replace = TRUE)),
       date = sample(seq(as.Date('2017/01/01'), as.Date('2018/08/31'), by="day"), 2000*7, replace = TRUE),
       ID = sample(1:2000, 2000*7, replace = TRUE),
       type = c(sample("blood", 1389*7, replace = TRUE),
                sample("other", 611*7, replace = TRUE))
) %>% 
  mutate(sample_type = ifelse(type == "blood", 
                              sample(blood_types, 1389*7, replace = TRUE), 
                              sample(other_types, 611*7, replace = TRUE))
    )

total_blood<-dfr_bio %>%
  count(type, ID, Site, Organ, sample_type) %>% 
  filter(type=="blood") %>%
  group_by(sample_type, Site) %>%
  summarise(samples=sum(n)) %>%
  spread(Site, samples, fill = 0)

total_others<-dfr_bio %>%
  count(type, ID, Site, Organ, sample_type) %>% 
  filter(type=="other") %>%
  group_by(sample_type, Site) %>%
  summarise(samples=sum(n)) %>%
  spread(Site, samples, fill = 0)

gen_bio_table<-function(organ, cat){
  df<-dfr_bio %>%
    count(type, ID, Site, Organ, sample_type) %>% 
    filter(Organ==organ & type==cat) %>%
    group_by(sample_type, Site) %>%
    summarise(samples=sum(n)) %>%
    spread(Site, samples, fill = 0)
  return(df)
}

organs_tab<-list("Kidney", "Liver", "Lung", "Heart", "Pancreas", "Stem Cells")
cat<-list("blood", "other")
tab_args<-expand.grid(organs_tab, cat)
names<-tab_args %>% unite(vector, c(Var1, Var2), sep = "_")

df_tab_bio<-NULL
for (i in 1:nrow(tab_args)){
  df_tab_bio[[i]]<-gen_bio_table(tab_args[i,1], tab_args[i,2])
}
names(df_tab_bio)<-names$vector
df_table <- df_tab_bio[[1]]
data <- df_tab_bio[[6]] %>% ungroup()

data <- NULL

gen_total <- function (data) {
  data <- data %>% ungroup()
  if (length(data)>1) {
    num_vecs <- c("Center_1", "Center_2", "Center_3", "Center_4", "Center_5")
    sum_data <- data %>% 
      rbind(c("TOTAL", c(select_if(data, is.numeric) %>% colSums())) %>% unlist() %>% as.vector()) %>% 
      mutate_at(vars(num_vecs), as.numeric) %>% 
      mutate(TOTAL = rowSums(.[2:6]))
    return(sum_data)
  }
}
 
df_final <- map(df_tab_bio, gen_total)
names(df_final) <- names$vector

total_blood <- gen_total(total_blood)
total_others <- gen_total(total_others)

blood_value <- total_blood %>%
  filter(sample_type=="TOTAL") %>%
  select(TOTAL) %>% as.numeric()

other_value <- total_others %>%
  filter(sample_type=="TOTAL") %>%
  select(TOTAL) %>% as.numeric()

pat_value <- dfr_bio %>%
  count(ID)
pat_value <- nrow(pat_value)


```


Column 
-----------------------------------------------------------------------

### Blutproben

```{r}
articles <- blood_value
valueBox(articles, icon = "ion-waterdrop", color = "#830303")
```

### Andere Proben

```{r}
comments <- other_value
valueBox(comments, icon = "fa-heart", color = "#B07C9E")
```

### Patienten mit Bioproben

```{r}
spam <- pat_value
valueBox(spam, icon = "ion-person-stalker", color = "#7EA16B")
```

Column {.tabset data-height=1300}
-----------------------------------------------------------------------
 **Blutproben**
 
### TOTAL

```{r}
datatable(
  total_blood, extensions = 'Buttons', options = list(
    dom = 'Bfrtip',
    buttons = c('copy', 'csv')
  )
)
```

### Niere
```{r}
datatable(
  df_final[[1]], extensions = 'Buttons', options = list(
    dom = 'Bfrtip',
    buttons = c('copy', 'csv')
  )
)
```

### Leber
```{r}
datatable(
  df_final[[2]], extensions = 'Buttons', options = list(
    dom = 'Bfrtip',
    buttons = c('copy', 'csv')
  )
)
```

### Lunge
```{r}
datatable(
  df_final[[3]], extensions = 'Buttons', options = list(
    dom = 'Bfrtip',
    buttons = c('copy', 'csv')
  )
)
```

### Herz
```{r}
datatable(
  df_final[[4]], extensions = 'Buttons', options = list(
    dom = 'Bfrtip',
    buttons = c('copy', 'csv')
  )
)
```

### Pankreas
```{r}
datatable(
  df_final[[5]], extensions = 'Buttons', options = list(
    dom = 'Bfrtip',
    buttons = c('copy', 'csv')
  )
)
```

### Stammzellen
```{r}
datatable(
  df_final[[6]], extensions = 'Buttons', options = list(
    dom = 'Bfrtip',
    buttons = c('copy', 'csv')
  )
)
```

Column {.tabset data-height=1000}
-----------------------------------------------------------------------
 **Andre Proben**


### TOTAL

```{r}
datatable(
  total_others, extensions = 'Buttons', options = list(
    dom = 'Bfrtip',
    buttons = c('copy', 'csv')
  )
)

```

### Niere
```{r}
datatable(
  df_final[[7]], extensions = 'Buttons', options = list(
    dom = 'Bfrtip',
    buttons = c('copy', 'csv')
  )
)

```

### Leber
```{r}
datatable(
  df_final[[8]], extensions = 'Buttons', options = list(
    dom = 'Bfrtip',
    buttons = c('copy', 'csv')
  )
)
```

### Lunge
```{r}
datatable(
  df_final[[9]], extensions = 'Buttons', options = list(
    dom = 'Bfrtip',
    buttons = c('copy', 'csv')
  )
)
```

### Herz
```{r}
datatable(
  df_final[[10]], extensions = 'Buttons', options = list(
    dom = 'Bfrtip',
    buttons = c('copy', 'csv')
  )
)
```

### Pankreas
```{r}
datatable(
  df_final[[11]], extensions = 'Buttons', options = list(
    dom = 'Bfrtip',
    buttons = c('copy', 'csv')
  )
)
```

### Stammzellen
```{r}
datatable(
  df_final[[12]], extensions = 'Buttons', options = list(
    dom = 'Bfrtip',
    buttons = c('copy', 'csv')
  )
)
```

Column {data-height=100}
-----------------------------------------------------------------------
[Doppelt klassifizierte Patienten](#page-3)


Page 3 {.hidden}
=======================================================================

### Doppelt klassifizierte Patienten
```{r eval=FALSE}
knitr::kable(doubles)
```

Follow - Up
=======================================================================

```{r FU_data_gen, echo=FALSE, warning=FALSE, message=FALSE}
# Follow-up generation
library(lubridate)
dfr_fu <- tibble(Site = sample(paste0("Center_", 1:5), 2000, replace = TRUE),
       Organ = c(sample("Heart", 58, replace = TRUE),
                 sample("Liver", 89, replace = TRUE),
                 sample("Lung", 47, replace = TRUE),
                 sample("Kidney", 871, replace = TRUE),
                 sample("Pancreas", 589, replace = TRUE),
                 sample("Stem Cells", 346, replace = TRUE)),
       TxDate = sample(seq(as.Date('2017/01/01'), as.Date('2018/08/31'), by="day"), 2000, replace = TRUE),
       FuDate = sample(seq(as.Date('2017/01/01'), as.Date('2018/08/31'), by="day"), 2000, replace = TRUE),
       ID = sample(1:2000, 2000, replace = TRUE)
) 

export_date <- dmy("31-8-2018")

fu_eval <- dfr_fu %>%
  mutate(TxPeriod=export_date-TxDate,
         FuPeriod=export_date-FuDate,
         Fu_3m=interval(TxDate %m+% months(3)-days(28), TxDate %m+% months(3)+days(28)), # when should the FU be?
         Fu_6m=interval(TxDate %m+% months(6)-days(28), TxDate %m+% months(6)+days(28)),
         Fu_9m=interval(TxDate %m+% months(9)-days(28), TxDate %m+% months(9)+days(28)),
         Fu_12m=interval(TxDate %m+% months(12)-days(28), TxDate %m+% months(12)+days(28)),
         checked_3m=FuDate %within% Fu_3m, # was the FU in this interval?
         checked_6m=FuDate %within% Fu_6m,
         checked_9m=FuDate %within% Fu_9m,
         checked_12m=FuDate %within% Fu_12m,
         mandatory_3m=Fu_3m %within% interval(TxDate,TxDate+TxPeriod), # should there be a FU in this interval?
         mandatory_6m=Fu_6m %within% interval(TxDate,TxDate+TxPeriod),
         mandatory_9m=Fu_9m %within% interval(TxDate,TxDate+TxPeriod),
         mandatory_12m=Fu_12m %within% interval(TxDate,TxDate+TxPeriod),
         eval_3m=ifelse(checked_3m==T & mandatory_3m==T, "ok",
                        ifelse(checked_3m==F & mandatory_3m==F, "ok",
                               ifelse(checked_3m==T & mandatory_3m==F, "not needed","not done"))),
         eval_6m=ifelse(checked_6m==T & mandatory_6m==T, "ok",
                        ifelse(checked_6m==F & mandatory_6m==F, "ok",
                               ifelse(checked_6m==T & mandatory_6m==F, "not needed","not done"))),
         eval_9m=ifelse(checked_9m==T & mandatory_9m==T, "ok",
                        ifelse(checked_9m==F & mandatory_9m==F, "ok",
                               ifelse(checked_9m==T & mandatory_9m==F, "not needed","not done"))),
         eval_12m=ifelse(checked_12m==T & mandatory_12m==T, "ok",
                         ifelse(checked_12m==F & mandatory_12m==F, "ok",
                                ifelse(checked_12m==T & mandatory_12m==F, "not needed","not done"))))

tab_fu <- fu_eval %>%
  select(ID, Organ, Site, TxDate, FuDate, contains("check"), contains("mandatory")) %>%
  filter(!(is.na(TxDate) | is.na(FuDate))) %>%
  select(-TxDate, -FuDate) %>%
  group_by(ID, Organ, Site) %>%
  summarise(ACTUAL_month_3=1*any(checked_3m),
            TARGET_month_3=1*any(mandatory_3m),
            ACTUAL_month_6=1*any(checked_6m),
            TARGET_month_6=1*any(mandatory_6m),
            ACTUAL_month_9=1*any(checked_9m),
            TARGET_month_9=1*any(mandatory_9m),
            ACTUAL_month_12=1*any(checked_12m),
            TARGET_month_12=1*any(mandatory_12m))

report_tab_fu <- tab_fu %>%
  group_by(Site, Organ) %>%
  summarise(ACTUAL_month_3=sum(ACTUAL_month_3),
            TARGET_month_3=sum(TARGET_month_3),
            Performance_month_3=round(100*ACTUAL_month_3/TARGET_month_3, digits=2),
            ACTUAL_month_6=sum(ACTUAL_month_6),
            TARGET_month_6=sum(TARGET_month_6),
            Performance_month_6=round(100*ACTUAL_month_6/TARGET_month_6, digits=2),
            ACTUAL_month_9=sum(ACTUAL_month_9),
            TARGET_month_9=sum(TARGET_month_9),
            Performance_month_9=round(100*ACTUAL_month_9/TARGET_month_9, digits=2),
            ACTUAL_month_12=sum(ACTUAL_month_12),
            TARGET_month_12=sum(TARGET_month_12),
            Performance_month_12=round(100*ACTUAL_month_12/TARGET_month_12, digits=2)) %>%
  ungroup()

#### which graph and tab has to be rendered
fu_exists <- report_tab_fu %>%
  group_by(Organ, Site) %>%
  summarise(count=n()) %>%
  select(-count) %>%
  ungroup() %>%
  as.data.frame()
fu_exists <- fu_exists %>% group_by(Organ) %>% summarise(count=n()) %>% as.data.frame()

sketch <- htmltools::withTags(table(class='display',thead(tr(
  th(rowspan=2, "Center"),
  th(colspan=3, "Month 3"),
  th(colspan=3, "Month 6"),
  th(colspan=3, "Month 9"),
  th(colspan=3, "Month 12")),
  tr(lapply(rep(c("ACTUAL", "TARGET", "in %"), 4), th)))))

tab <- function(organ){
  tab <- report_tab_fu %>%
    filter(Organ==organ) %>%
    select(-Organ)
  tab <- datatable(tab, container = sketch, rownames=F)
  return(tab)
}

tables <- NULL
for (i in 1:nrow(fu_exists)) {
  tables[[i]] <- tab(fu_exists[i,1])
}

#### Figure Preparation            
fig_fu <- report_tab_fu %>%
  gather(contains("ACTU"), contains("TARG"), key="time", value = "Patients") %>%
  select(-contains("Perf")) %>%
  separate(time, c("ind", "times", "month")) %>% 
  select(-contains("time"))
fig_fu$month <- sapply(fig_fu$month, as.numeric)
fig_fu <- fig_fu %>% select(Site, Organ, Indikator=ind, Monat=month, Patienten=Patients)

graph <- function(organ){
  g <- fig_fu %>%
    filter(Organ==organ) %>%
    arrange(desc(Indikator))
  g$Indikator <- factor(g$Indikator, ordered = T)
  g$Indikator <- factor(g$Indikator,levels(g$Indikator)[c(2,1)])
  g <- g %>%
    ggplot(aes(x=Monat,y=Patienten,group=Indikator,color=Indikator,fill=Indikator)) +
    geom_line() +
    geom_area(position="identity") +
    facet_wrap("Site",nrow=1) +
    scale_color_manual(values=c("#CC6666", "#66CC99"))+
    scale_fill_manual(values=c("#CC6666", "#66CC99")) +
    scale_x_continuous(breaks=c(3,6,9,12)) +
    #scale_y_continuous(position="right") +
    #theme(axis.text.x = element_text(angle = 45)) +
    theme_bw()
  #theme(strip.text = element_text(size=16), 
  #      axis.text.x = element_text(angle = 45, size=12), 
  #      axis.text.y = element_text(size=12))
  g <- ggplotly(g, height=300, width=1500)
  
  return(g)
}

graphs <- NULL
for (i in 1:nrow(fu_exists)) {
  graphs[[i]] <- graph(fu_exists[i,1])
}

#### Save 4 RMD - TABs & GRAPHs ####
fu_herz <- list(tables[[1]], graphs[[1]])
fu_leber <- list(tables[[2]], graphs[[2]])
fu_lunge <- list(tables[[3]], graphs[[3]])
fu_niere <- list(tables[[4]], graphs[[4]])
fu_pankreas <- list(tables[[5]], graphs[[5]])
fu_stammzellen <- list(tables[[6]], graphs[[6]])


```



Row {.tabset .tabset-fade} 
-----------------------------------------------------------------------

### Übersicht Niere FU
Anzahl der Patienten im FU
```{r}
fu_niere[[2]]
```

### Übersicht Leber FU
Anzahl der Patienten im FU
```{r}
fu_leber[[2]]
```

### Übersicht Lunge FU
Anzahl der Patienten im FU
```{r}
fu_lunge[[2]]
```

### Übersicht Pankreas FU
Anzahl der Patienten im FU
```{r}
fu_pankreas[[2]]
```

### Übersicht Stammzellen FU
Anzahl der Patienten im FU
```{r}
fu_stammzellen[[2]]
```

### Übersicht Herz FU
Anzahl der Patienten im FU
```{r}
fu_herz[[2]]
```

Row {.tabset .tabset-fade} 
-----------------------------------------------------------------------
### Übersicht Niere FU
```{r}
fu_niere[[1]]
```

### Übersicht Leber FU
```{r}
fu_leber[[1]]
```

### Übersicht Lunge FU
```{r}
fu_lunge[[1]]
```

### Übersicht Pankreas FU
```{r}
fu_pankreas[[1]]
```

### Übersicht Stammzellen FU
```{r}
fu_stammzellen[[1]]
```

### Übersicht Herz FU
```{r}
fu_herz[[1]]
```

Row {data-height=100}
-----------------------------------------------------------------------
### Daten - Probleme
[Übersicht Patienten ohne Tx und FU Datum](#page-4)


Page 4 {.hidden}
=======================================================================

### Übersicht Patienten ohne Tx und FU Datum
```{r eval=FALSE}
no_TxDate
no_FuDate
```


Generell {data-navmenu="Datenqualität" data-icon="fa-list" data-orientation=columns}
=============================

```{r dq_data_gen, echo=FALSE, warning=FALSE, message=FALSE}
### generate data quality
simulate_table <- function() {
  tibble(Variable = sample(paste0("Item_", 1:9), 9),
         Center_1 = sample(75:90, 9, replace = TRUE),
         Center_2 = sample(75:90, 9, replace = TRUE),
         Center_3 = sample(75:90, 9, replace = TRUE),
         Center_4 = sample(75:90, 9, replace = TRUE),
         Center_5 = sample(75:90, 9, replace = TRUE))
}
gen_total2 <- function (data) {
  if (length(data)>1) {
    num_vecs <- c("Center_1", "Center_2", "Center_3", "Center_4", "Center_5")
    sum_data <- data %>% 
      rbind(c("TOTAL", c(select_if(data, is.numeric) %>% colMeans())) %>% unlist() %>% as.vector()) %>% 
      mutate_at(vars(num_vecs), as.numeric) %>% 
      mutate(TOTAL = rowMeans(.[2:6])) %>% 
      mutate_if(is.numeric, round, digits = 0)
    return(sum_data)
  }
}
col_rank <- function (vec) {
  # function to color background in accordance to its value
  # input vector : vec
  # output vector : fin
  farben <- length(unique(vec))
  palette <- brewer.pal(farben, "RdYlBu")
  max_farben <- vec %>% as_tibble %>% 
    arrange(value) %>% distinct(value) %>% 
    cbind(palette) 
  fin <- vec %>% as_tibble() %>% 
    left_join(max_farben) %>% # here we join the unique colors with its values
    select(2) %>% unlist() %>% as.vector()
  return(fin)
}
make_tbl_colorful <- function (table) {
  gen_total2(table) %>% 
    arrange(Variable) %>%
    mutate_if(is.numeric, function (x) cell_spec(x, background = col_rank(x), color = "#a8a8a8", bold = TRUE)) %>% 
    kable(escape = F, align = "c") %>%
    kable_styling(c("striped", "condensed"), full_width = F)
}

make_table <- function(){
  simulate_table() %>% 
    gen_total2() %>% 
    make_tbl_colorful()
}

tab_1oc <- simulate_table() %>% gen_total2()
tab1 <- make_tbl_colorful(tab_1oc)

tab_2oc <- simulate_table() %>% gen_total2()
tab2 <- make_tbl_colorful(tab_2oc)

tab_3oc <- simulate_table() %>% gen_total2()
tab3 <- make_tbl_colorful(tab_3oc)

ranking_gen <- tab_1oc %>% filter(grepl("TOTAL", Variable)) %>% select(-1, -TOTAL) %>% unlist() 
ranking_postop <- tab_1oc %>% filter(grepl("TOTAL", Variable)) %>% select(-1, -TOTAL) %>% unlist() 
ranking_bak <- tab_1oc %>% filter(grepl("TOTAL", Variable)) %>% select(-1, -TOTAL) %>% unlist() 

avg_gen_site <- tab_1oc %>% filter(grepl("TOTAL", Variable)) %>% select(TOTAL) %>% unlist() %>% as.vector()
avg_postop <- tab_2oc %>% filter(grepl("TOTAL", Variable)) %>% select(TOTAL) %>% unlist() %>% as.vector()
avg_bak <- tab_3oc %>% filter(grepl("TOTAL", Variable)) %>% select(TOTAL) %>% unlist() %>% as.vector()

make_ranking_trend <- function(datum) {
  id <- ymd(datum)
  simulate_table() %>% gen_total2() %>% filter(grepl("TOTAL", Variable)) %>% 
    select(-1, -TOTAL) %>% 
    mutate(id = id)
}

datum <- sapply(seq(ymd('2017-1-1'), ymd('2018-8-1'), by = "month"), list)

gen_plot <- map_dfr(datum, make_ranking_trend)
postop_plot <- map_dfr(datum, make_ranking_trend)
bak_plot <- map_dfr(datum, make_ranking_trend)



```



Column {data-width=200}
-----------------------------------------------------------------------

### Datenqualitaet Generelle Informationen
```{r}
rate <- avg_gen_site
gauge(rate, min = 0, max = 100, symbol = '%', gaugeSectors(
  success = c(80, 100), warning = c(60, 79.99), danger = c(0, 59.99)
))
```

### Ranking
```{r}
knitr::kable(ranking_gen, "html") %>%
  kable_styling("striped", full_width = F) %>%
  row_spec(1:3, bold = T, font_size=18, color = "white", background = "#4292c6")
```

### Datenqualitaet ueber Zeit
```{r}
g <- gen_plot %>% 
  gather(center, missing, -id) %>% 
  ggplot(aes(id, missing, color = center)) +
  geom_line() +
  scale_color_viridis_d()

plotly::ggplotly(g)
```


Column
-----------------------------------------------------------------------

### Variable pro Center

```{r}
tab1
```


Post-op {data-navmenu="Datenqualität" data-icon="fa-list" data-orientation=columns}
=============================

Column {data-width=200}
-----------------------------------------------------------------------

### Datenqualitaet Post operative Informationen
```{r}
rate <- avg_postop
gauge(rate, min = 0, max = 100, symbol = '%', gaugeSectors(
  success = c(80, 100), warning = c(60, 79.99), danger = c(0, 59.99)
))
```

### Ranking
```{r}
knitr::kable(ranking_postop, "html") %>%
  kable_styling("striped", full_width = F) %>%
  row_spec(1:3, bold = T, font_size=18, color = "white", background = "#4292c6")
```

### Datenqualitaet ueber Zeit
```{r}
g2 <- postop_plot %>% 
  gather(center, missing, -id) %>% 
  ggplot(aes(id, missing, color = center)) +
  geom_line() +
  scale_color_viridis_d()

plotly::ggplotly(g2)

```

Column
-----------------------------------------------------------------------

### Variable pro Center

```{r}
tab2
```

Bakterielle Infektionen {data-navmenu="Datenqualität" data-icon="fa-list" data-orientation=columns}
=============================

Column {data-width=200}
-----------------------------------------------------------------------

### Datenqualitaet Bakterielle Infektionenn
```{r}
rate <- avg_bak
gauge(rate, min = 0, max = 100, symbol = '%', gaugeSectors(
  success = c(80, 100), warning = c(60, 79.99), danger = c(0, 59.99)
))
```

### Ranking
```{r}
knitr::kable(ranking_bak, "html") %>%
  kable_styling("striped", full_width = F) %>%
  row_spec(1:3, bold = T, font_size=18, color = "white", background = "#4292c6")
```

### Datenqualitaet ueber Zeit
```{r}
g3 <- bak_plot %>% 
  gather(center, complete, -id) %>% 
  ggplot(aes(id, complete, color = center)) +
  geom_line() +
  scale_color_viridis_d()

plotly::ggplotly(g3)

```

Column
-----------------------------------------------------------------------

### Variable pro Center

```{r}
tab3
```